/tools/ab-test-significance-calculator

Two gauges. One verdict.

Frequentist on the left. Bayesian on the right. Punch in the numbers and read what they actually say — without writing a paper on multiple comparisons first.

Variant A — visitors

A — conversions

Variant B — visitors

B — conversions

Frequentist

Two-proportion z-test

Confidence

—

Awaiting input

p-value

—

z-score

—

95% confidenceNot met

99% confidenceNot met

WinnerInconclusive

Bayesian

Beta–Binomial, uniform prior

P(B beats A)

—

Awaiting input

P(B > A)

—

E[loss | ship B]

—

95% probNot met

99% probNot met

Recommendation—

How to read this result · Awaiting data

Enter all four numbers above.

Both gauges activate the moment you give them visitors and conversions for A and B. Nothing happens until you do.

Generates a verdict-card preview when you drop the link in Slack, Linear, or anywhere with link unfurling.